A clinical governance lead for a regional imaging network receives the quarterly compliance summary. Ninety-four percent of radiographers across the group hold current modality credentials. The figure is, in the terms the report uses, good. It is above the threshold. It will appear in the board report as a positive indicator. The meeting moves on.
Later, at her desk, she tries to answer a question that the summary did not address: which six percent? Not a rounded cohort figure — the specific radiographers, the specific modalities, the specific nature of the gap. Is the six percent clustered at one site, or distributed? Are the outstanding verifications a matter of administrative timing — renewals in process — or do they represent genuine lapses in standing? Are any of the unverified radiographers currently rostered? If so, in what roles, covering which modalities, at which sites?
The compliance summary cannot answer any of these questions. It was not designed to. It was designed to produce the headline figure, and it has produced it accurately. The gap between the headline figure and the questions that responsible governance actually requires is not a reporting failure. It is a resolution failure — the governance data exists at a level of aggregation that makes individual accountability invisible.
The aggregation convention
This is the third and final feature of continuous governance that distinguishes it from audit-oriented practice. The first article in this series identified the structural gap between audit-readiness and being genuinely well-governed. The second argued that currency — data that is queryable in present tense — is the property that makes governance information operationally useful. The third, the proximate governance signal, established that being well-governed at the institutional level is not equivalent to being well-governed at the point of clinical activity. This article addresses the last feature: the resolution at which governance data operates, and what happens when that resolution is too coarse to support the decisions governance is supposed to enable.
Aggregate compliance figures are a convention of convenience. They compress practitioner-level data into a departmental or organisational metric because that is the grain at which boards and regulators have typically asked the question. Ninety-four percent is a comprehensible figure. It appears in a report, it invites comparison against a benchmark, it supports a governance narrative. What it cannot do is identify whether the specific radiographers who are currently not credential-current are working today, in what contexts, and with what specific gaps in their standing.
The convention persists in part because the underlying data has not, historically, been available at sufficient granularity to support a more precise question. Competency verification has been conducted periodically and recorded in forms designed to produce aggregate outputs. The individual practitioner record, where it exists in detail, is held in HR systems that were not designed to be queried operationally, or in departmental records that are maintained locally and consulted manually when a specific question arises. The aggregate figure is the grain at which governance information has reliably existed. It is not the grain at which governance accountability operates.
The practitioner-level question
Accountability granularity, as a governance property, means that the system operates at the level of the individual practitioner and the individual episode, where that precision is clinically meaningful. The emphasis on "where clinically meaningful" is deliberate. Not every governance question requires individual-level resolution. Workforce planning questions — how many MRI sessions can the department support next quarter — are legitimately answered at the aggregate level. The question of whether a specific radiographer's current standing covers the specific activity they are about to undertake is not.
The difference between "eligible" and "ineligible for a named, specific reason" is a category difference in governance accountability, not merely a difference of detail. Consider what accountability granularity makes possible. A governance system that returns a binary result — this radiographer is eligible, or is not — provides a decision gate. A governance system that returns the specific reasons a radiographer's standing is not confirmed — whether it concerns the recency of modality-specific independent practice, the currency of MR safety training, the status of reference confirmation, the standing of consent and pool membership, or any of several other named, dated, modality-specific standards — provides a diagnostic signal.
The diagnostic signal does something the binary gate cannot. It tells the governance lead not only that a gap exists but what kind of gap it is, which determines what should be done about it. An MR safety training currency issue has a different resolution pathway from a modality recency question, which has a different pathway again from an outstanding reference confirmation. The aggregate compliance figure tells you how many gaps there are. The granular signal tells you what they are, which is the only basis on which they can be specifically addressed rather than generically managed.
What becomes visible
When accountability granularity is present, the six percent becomes answerable. Not as a matter of interest, but as a matter of governance function. The clinical governance lead can determine not only which radiographers are not currently confirmed eligible for specific modalities, but the named category of the reason — and therefore assess whether those radiographers are appropriately rostered in their current positions, whether the gap is administrative or substantive, and what the timeline and process for resolution looks like. This is not surveillance. It is the basic operating requirement of a governance system that is accountable at the level at which clinical risk exists.
The patient-safety argument for granularity is not, it should be said, primarily about finding radiographers who should not be working and are. It is about the structural condition in which that question can never be cleanly answered. In a governance system that operates only at aggregate resolution, the absence of evidence for an individual-level problem is not evidence of its absence — it is a consequence of the grain at which the system operates. The six percent remains a cohort. The radiographers within it remain unidentified. The reasons they are within it remain unknown. The governance narrative says "good." The clinical reality may or may not align.
What this argument does not claim
Individual-level governance data does not make governance decisions automatically. The granular signal still requires a person — a clinical governance lead, a service manager, a department head — to review it, interpret it, and act on it. What changes is the quality and specificity of what that person is acting on. The difference between managing a cohort label and managing a named gap in a named radiographer's standing for a named modality is the difference between governance as a reporting exercise and governance as an operational function.
It is also worth saying plainly that a governance system operating at practitioner-level granularity is only as useful as the underlying data is current and proximate. Granularity without currency produces a highly detailed picture of a state of affairs that no longer exists. Granularity without proximity produces precise information about credentials held in one context that may not apply to another. The three features are interdependent; this series has tried to take them in the order that makes each argument legible on its own terms.
Remote-I's eligibility architecture returns, for any radiographer-modality combination in a hospital's pool, the specific named reasons a radiographer's standing is confirmed or not — drawing on categories that span consent and pool status, experience floors for total years and modality-specific independent practice, modality recency, MR safety training currency, reference standing, and governance approval status. The result is not a score, and it is not a binary flag. It is a current, named, modality-specific accountability record. That is what governance at adequate resolution looks like, and it is a different category of thing from the quarterly compliance summary.
A governance system that can only tell you what percentage are compliant has, in effect, decided in advance that the individual-level question will not be asked. The more considered decision is to build a system in which it can be.